Nowadays, in general, a large amount of data are stored in a server for facilitating a client to download the data to its terminal device, so as to desirably reduce costs of data maintenance, renewal and transmission. Besides, for obtaining the latest information in real time, terminal devices of clients can request for downloading updated data stored in the server.
However, the foregoing client/server system is disadvantageous in that, if too many clients request for data downloading from the server at the same time, overload of the server and reduction in available network bandwidth can easily occur; this in turn impairs the proper operation of network. Moreover, as to prevent the over occupation of network bandwidth for allowing the smooth operation of network, if the frequency of data-retrieval requests from the clients to the server is limited, or time intervals of request submission between the clients are prolonged, then the clients would be unable to obtain data in real time; in the case of more clients in operation, data retrieval efficiency becomes even worse.
Therefore, in accordance with the foregoing drawbacks, how to reduce the server's load and occupation of network bandwidth for providing an efficient real time data transmission, is a critical problem to solve.